This research explores the operation of perceived self-efficacy in learning-disabled children during cognitive skill development. Within this context, the major issues are: (a) Does self-verbalization during skill acquisition promote self-efficacy and skills, and(b) Does the sequencing of effort attributional feedback differentially affect these outcomes? Children who have been classified as learning-disabled in mathematics will participate in a subtraction competency-development program over multiple sessions, during which they will receive cognitive modeling of operations and opportunities to solve problems. During the problem solving, one group of children will verbalize aloud solution steps and their application, a second group will verbalize initially but then will fade verbalizations to a covert level, and a third group will not verbalize. Children will be monitored periodically by an adult proctor who will deliver effort attributional feedback that links their progress with effort. Some children will receive effort feedback during the entire training program, others will receive it only during the first half of the program, and a third group of children will be given it only during the second half. Dependent measures include self-efficacy, persistence, skill, and attributions. This research assumes that self-efficacy constitutes an important variable in understanding learning-disabled children's self-functioning. Even though such children possess the ability to learn, they often view cognitive tasks as insurmountable, which results in behavioral dysfunctions. A sense of inefficacy for coping with cognitive demands often is associated with adverse emotional reactions to stressful situations, which can promote self-doubts, result in lackadaisical efforts, and preclude skill development. Positive results from this research would suggest developing a comprehensive coping skills training program and its extension to other child populations with congnitive skill deficits.